Because each of the position transducers commonly
used in industry, i.e., encoders and resolvers, does not inherently
measure an instantaneous velocity, some signal processing technique
is generally required to improve the accuracy of velocity estimation
at each sampling instant. This estimated signal is then used
as the velocity feedback signal for the velocity loop control. The differential
position technique commonly used in industry for velocity
estimation suffers from a large quantization error and, therefore,
limits the servo control loop bandwidth. This paper presents both
the theoretical analysis and experimental verification of a scheme
that uses a closed-loop observer for velocity estimation. The results
have shown that the quantization error in the velocity feedback
signal can be reduced dramatically when a closed-loop observer
is used for velocity estimation. The results also shown that, at
low speed, the velocity estimation could be improved with a simple
compensation scheme.